-1372,6 +1374,7 @@
> */
> if( ORTE_SUCCESS != (ret =
> orte_errmgr_base_restart_job(current_global_jobid, global_handle, seq_num) )
> ) {
> ORTE_ERROR_LOG(ret);
> +orte_snapc_ckpt_state_notify(ORTE_SNAPC_CKPT_STATE_NO_RESTART
I was going through the MCA btl parameters for openib and came across this
parameter in which i didn't get what it actually does:
MCA btl: parameter "btl_openib_exclusivity" (current value: "1024")
BTL exclusivity (must be >= 0)
anybody
r turns off
>> sparse groups, or
>> b) there's an added check in the inline function for whether the MCA
>> parameter is on
>>
>> I'm personally in favor of a) because it means only one conditional
>> in the critical path.
>
> I don't really c
Sorry the additional parameters are 5 integers and 3 pointers..
Mohamad
On Wed, July 25, 2007 9:45 pm, Mohamad Chaarawi wrote:
> In the current code, almost all #ifs are due to the fact that we don't
> want to add the extra memory by the sparse parameters that are added to
> the g
ode, which im
for..
>
> Other than that, I think you've got my concerns pretty much addressed.
>
> Brian
>
> On Jul 25, 2007, at 8:45 PM, Mohamad Chaarawi wrote:
>
>> In the current code, almost all #ifs are due to the fact that we don't
>> want to a
On Thu, July 26, 2007 1:18 pm, Brian Barrett wrote:
> On Jul 26, 2007, at 12:00 PM, Mohamad Chaarawi wrote:
>
>>> 2) I think it would be better to always have the flags and macros
>>> available (like OMPI_GROUP_SPORADIC and OMPI_GROUP_IS_SPORADIC) even
>>> when
On Thu, July 26, 2007 2:07 pm, Brian Barrett wrote:
> On Jul 26, 2007, at 1:01 PM, Mohamad Chaarawi wrote:
>
>>
>> On Thu, July 26, 2007 1:18 pm, Brian Barrett wrote:
>>> On Jul 26, 2007, at 12:00 PM, Mohamad Chaarawi wrote:
>>>
>>>>> 2) I th
al.
>
Another proposition that i mentioned before is to keep the sparse
parameters in the group structure (not compile them out) when the sparse
groups are disabled, which will remove almost all #ifs from the code,
which will be much easier for the eyes (the main reason).. Brian had some
object
need to add a line to ompi_info to indicate wether the sparse
groups are indicated in the build or not.
I will most probably commit tomorrow afternoon or sometime in the weekend,
Unless someone objects..
Thank you,
--
Mohamad Chaarawi
Instructional Assistant http://www.cs.uh.edu
proc_count) {
> +if (peer_id >= group->grp_proc_count || peer_id < 0) {
> opal_output(0, "ompi_group_lookup_peer: invalid peer index
> (%d)", peer_id);
>
>> Thanks,
>>
>> Tim
>> ___
>>
e time.
>
> But if you apply the patch I sent, it will tell you when an invaild
> lookup happens, which should be every time it runs.
>
> Tim
>
> Mohamad Chaarawi wrote:
>> Hey Tim,
>>
>> I understand what you are talking about.
>> Im trying to reproduce t
11 matches
Mail list logo